The present invention relates generally to locating an image to be scanned and, in particular, to processing data from an initial scan to determine an area including the image and scanning the image a second time from the determined area.
A flatbed scanner is a device that converts an image into digital data values. In conventional flatbed scanner systems, an image, such as a photograph or document, is placed on a bed of the scanner, and the scanner scans the bed, including the image, and converts the scanned area into an array of data values. Each data value corresponds to a different area of the scanned bed. Two factors determine how much data is generated by the scan: the resolution of the scan and the type of information generated by the scan.
The first factor, resolution, is based on the size of each area from which data is generated. For example, one inch by one inch areas would yield a lower resolution than areas of 0.1 inch by 0.1 inch because higher resolutions have more data values for each area. Resolution is often measured in dots per inch (dpi). A low resolution would be 50 dpi, while a high resolution for certain systems is 600 dpi.
The second factor, the type of information generated by the scan, is based on the amount of information calculated for each area, which typically depends on whether the scan is for black and white, gray scale, or color. For example, if each area is merely converted to black and white colors, then each area only can be represented by a single bit. On the other hand, if each area is converted to color, each area must be represented by more bits, for example, eight bits for red, eight bits for green, and eight bits for blue.
In addition to the resolution and the type of information generated by the scan, the size of the scanned area also affects the amount of data from the scan. In conventional systems the entire bed is scanned, even though the image may only occupy a fraction of the scanner bed. Techniques have been developed for determining an area containing the image so the scanner need only scan that area, but such techniques are often not accurate because they do not account for characteristics of different types of images. For example, when scanning a text image, conventional systems often scan the bottom line of text in a manner that misses underscores because conventional scanning techniques assume that the image starts at the lowermost edge of the last line of text, which is above an underscore.
Conventional systems also scan every image the same way, irrespective of the image color. For example, most systems simply assume the image is color, and scan the image to develop red, green and blue values. This creates the same amount of data irrespective of whether the data is black and white or color.
Apparatus and methods consistent with this invention provide for an improved determination of image characteristics. An apparatus consistent with the principles of the invention for determining an image area in a scanning area containing an image comprises an element for receiving initial scan data corresponding to the scanning area, an element for processing the initial scan data to categorize the image by type, and an element for determining, based on the type, the image area which is a subset of the scanning area and includes the image. Because the apparatus determines the type of image, and uses the type of image in determining a subset of the scanning area, a more accurate scan can be performed.
A method consistent with the principles of the invention comprises receiving initial scan data corresponding to the scanning area, processing the initial scan data to categorize the image by type, and determining, based on the type, the image area which is a subset of the scanning area and includes the image. Similar to the apparatus, because the method determines the type of image, and uses the type of image in determining a subset of the scanning area, a more accurate scan can be performed.
The invention may also be embodied in a system, an computer-readable medium, a signal, or other form. Each form provides an improved way of scanning an image area.